On-Chain Credit Protocols: Default History and Lender Returns

By Chanuka Geekiyanage Oct 27, 2026

On-chain credit protocols pay lenders 8% to 17% because lenders carry real borrower risk, and the default record shows how often that risk has turned into losses. Maple, Goldfinch, Centrifuge, TrueFi and Clearpool have all had borrower failures since 2022, and the outcomes differed sharply by loss structure. This article reviews the documented defaults, shows what they did to lender returns, and gives a checklist for deciding whether a given pool is worth your capital. Figures are as of late August to September 2026 unless stated, and yields change often.

Panaprium ist unabhängig und wird vom Leser unterstützt. Wenn Sie über unseren Link etwas kaufen, erhalten wir möglicherweise eine Provision. Wenn Sie können, unterstützen Sie uns bitte monatlich. Die Einrichtung dauert weniger als eine Minute und Sie werden jeden Monat einen großen Beitrag leisten. Danke schön!

The Default Record, Protocol by Protocol

Maple Finance: One Borrower, One-Third of the Book

In December 2022, Orthogonal Trading defaulted on eight loans totalling $36 million, about 30% of active loans across Maple. About $31 million sat in the M11 USDC pool, which took roughly an 80% loss, while the Maven 11 WETH pool lost 17%. The borrower had misstated its FTX exposure, and Maple said Orthogonal’s share of the M11 book was 14% as of September 1, within the pool’s limits.

The pool had looked healthy. By October 2022 it had paid lenders an average rate of 10.76%. At that rate, an 80% loss takes about 7.4 years of simple interest to earn back. Concentration, not the headline rate, decided the outcome.

Maple then rebuilt its loss rules. Maple 2.0 requires pool delegates to post first-loss capital, and defaulted loans are written down immediately while recovery continues. Lenders can also request withdrawals at any time instead of waiting out a 30-day lock-up. Today the flagship products are overcollateralized: in late August 2026 syrupUSDC paid 4.8% APY and syrupUSDT 4.3%, with Maple reporting $4.82B in AUM against $2.918B in TVL on DefiLlama. The gap is a measurement difference, so never compare one protocol’s AUM with another’s TVL. In the sources I reviewed I found no reported pool-wide writedown since the 2023 relaunch, but that is absence of evidence, so check Maple’s own transparency dashboard.

Our breakdown of Maple, Goldfinch and Centrifuge covers how the three differ at the product level.

Image source: app.rwa.xyz/credit

Goldfinch: Emerging-Market Loans That Did Not Repay

Goldfinch had three major credit events. Tugende Kenya’s $5 million loan defaulted in 2023, Stratos, a US credit fund, failed on a $7 million portion, and Lend East could repay only about $4.25 million of $10.2 million. Warbler Labs, the core developer, backstopped the Stratos losses, but nobody backstopped Lend East.

The Lend East pool shows the return math clearly. It offered 17% USDC APY on a 25-month term, and the shortfall was 58% of principal, about 7.7% of Goldfinch’s active loans. If a lender had received all interest in full over the term (roughly 35% of principal), the net result would still be about 24% below the starting capital. The exact result depends on payment timing and recoveries, so treat that as a rough estimate.

The structure also failed in a way users did not expect. The senior pool was marketed as protected by first-loss backers, but once defaults exceeded the junior layer, senior participants absorbed losses. In June 2026 a governance vote closed June 23 approved winding down Goldfinch Prime and moving the protocol to maintenance mode, with about $56 million of borrower capital reported frozen. A depositor alleged that more than $50 million in loans across eight borrowers were either in default or restructuring. That figure is a depositor’s claim, not an audited number.

Centrifuge: Tranches First, Then a Pivot

Centrifuge’s original Tinlake design used tranches. In a default, the junior tranche takes losses first and earns a higher yield, while the senior tranche earns less. Senior investors are only affected once defaults exceed the pool’s junior ratio. That design held better than Goldfinch’s, but it still recorded losses: an August 2023 rwa.xyz snapshot showed defaulted loans of about $48.2 million at Maple, $13.2 million at Centrifuge and $4.4 million at TrueFi. That snapshot is secondary and dated, so verify it on rwa.xyz.

The product mix has changed. Two Janus Henderson funds make up about 95.6% of the platform, and TVL is around $1.6 billion after peaking near $2 billion earlier in 2026. That is institutional fund exposure (Treasuries and AAA CLOs), not small-business loan risk. Lower credit risk also means lower yield.

TrueFi and Clearpool: Unsecured Lending Without Much Cushion

TrueFi’s 2022 losses came from trading-firm borrowers. Alameda represented half of outstanding loans, and default protection was about $1.36 million against roughly $4 million of bad debt. That mechanism slashes staked TRU tokens, so it is only as strong as the token’s price.

Clearpool has a different weakness. The smart contracts do no underwriting and rely on third-party scoring from Credora. At 99% utilization a pool enters a Warning state where neither side can withdraw. In the FTX panic, Amber Group, Auros and LedgerPrime hit that warning, and borrowing costs spiked from 8-10% to 20-25% APR. Lenders who wanted out could not leave.

Stream Finance in 2025: Credit Risk Through Collateral

Not every credit loss comes from a borrower in a credit pool. In November 2025, Stream Finance’s debt to DeFi lenders was estimated at $285 million after the issuer reported a $93 million fund loss. Curated lending vaults on Morpho, Euler and Silo held xUSD-related positions. MEV Capital realized 3.6% bad debt in its Ethereum USDC vault and about $600,000 of loss on roughly $5 million remaining in its Arbitrum vault. Vault depositors paid for a curator’s collateral choice.

Comparing the Documented Outcomes

Protocol	Documented credit event	Loss structure	Current status	Best for
Maple	Orthogonal, Dec 2022 (~$36M)	Delegate first-loss capital after 2.0; immediate write-downs	Operating, ~$4.8B AUM	Lenders accepting institutional credit risk
Goldfinch	Tugende, Stratos, Lend East	Backers first loss; senior pool leveraged	Wind-down approved June 2026	Nobody new; recovery only
Centrifuge	Tinlake pool defaults	Junior/senior tranches (legacy)	~$1.6B, mostly Janus Henderson funds	Treasury and AAA-CLO exposure
TrueFi	Alameda, Blockwater, others	TRU-staker default protection	Far smaller than 2022 peak	Specialists only
Clearpool	Utilization stress, Nov 2022	Insurance accounts, auction on default	Operating, reduced scale	Lenders who accept lockup risk

What Defaults Do to Lender Returns

A simple break-even test helps. A pool yielding 12% with a 60% loss on default beats a 4.8% alternative only if annual default odds stay below roughly 10% (12 minus 4.8, divided by 12 plus 60). That sounds easy until one borrower is 30% of the pool, as at Maple in 2022. Then a single default does more damage than years of interest can offset.

Market size can also mislead. rwa.xyz showed about $12.9 billion in active private-credit loans and a 10.16% average APR, but Figure alone accounted for over $9.5 billion of that. Most of the headline is not lender-facing DeFi credit, so it says little about what a stablecoin depositor earns.

Image source: defillama.com/protocol/maple-finance

How to Check a Credit Pool Before Depositing

Before committing capital, check these items:

Borrower concentration: Find the largest borrower’s share. Maple’s M11 pool had Wintermute near 42% and Orthogonal near 26% of originations (per BlockCrunch), and one failure still cost 80%.
Who absorbs first loss: Look for real capital from the underwriter or delegate, not just tokens that can fall to zero.
Collateral: Overcollateralized loans change the default math. Unsecured loans need much higher yield to justify the same risk.
Withdrawal rules: Check lockups, queue mechanics and utilization caps. Clearpool’s Warning state froze lenders when they most needed an exit.
Where the yield comes from: Know the borrower type and whether token incentives inflate the APY.
Disclosure record: Read the protocol’s governance forum for workouts and missed payments, not just the dashboard.

On-chain data tools help here, and our guide on using AI to analyze on-chain data safely explains how to verify what an AI tool tells you before acting on it.

Common Mistakes

Trusting the senior label. Goldfinch’s senior pool absorbed losses once the junior layer ran out.
Chasing the 17%. A high APY on a single-borrower pool is a loss estimate, not a bonus.
Ignoring composability. Yield-bearing credit tokens used as collateral elsewhere can create exit pressure in stress.
Reading AUM as safety. Size does not measure underwriting quality.
Ignoring curator risk. Stream showed that vault curators can import hidden credit exposure.

Best Choice by Lender Type

Lender type	Recommended option	Why
Wants low credit risk, accepts modest yield	Centrifuge tokenized funds (JAAA or JTRSY)	Institutional fund assets, not single-borrower loans
Wants higher yield with institutional underwriting	Maple syrup products	Overcollateralized loans and post-2022 loss rules
Experienced, wants unsecured exposure	Small, capped TrueFi or Clearpool positions	Higher yield, but weak first-loss and lockup risk
New lender	Avoid Goldfinch	Wound down and in recovery

My Take

For most lenders, Maple’s overcollateralized products are the only credit option with a clear loss structure, a public track record since the 2023 relaunch and meaningful scale. The yield is modest, 4.3% to 4.8% in August 2026, which is the price of removing the single-borrower risk that broke the 2022 pools. If you want higher yield, size the position as venture risk, not savings.

Avoid anything where one borrower exceeds roughly 15% of a pool, where first-loss capital is only a token, or where withdrawals can freeze at high utilization. Before depositing I would check the loan book, the largest borrower’s share, the withdrawal queue and the latest governance forum posts. Centrifuge’s fund wrappers are better for lower-risk exposure, but they are really tokenized Treasury and CLO products, so do not compare their yield with Maple’s.

Conclusion

The default record points to one conclusion: lender returns depend more on loss structure and concentration than on the advertised APY. Maple recovered by adding first-loss capital and overcollateralization, Goldfinch’s senior pool absorbed losses it was never meant to take, and Stream showed that vault curators can import credit risk without lenders seeing it.

Your next step is to pick one pool, read its loan book and governance forum, and calculate a break-even default rate against a lower-risk alternative. If you cannot find the largest borrower’s share or the first-loss capital, treat that as a reason to skip it.

FAQs

1. Has Maple Finance had defaults since 2022?

I found no reported pool-wide writedown since the 2023 relaunch in the sources I reviewed. Verify against Maple’s transparency dashboard before depositing, because that finding is an absence of evidence.

2. Is Goldfinch still safe to use?

No, because Goldfinch Prime is being wound down and the protocol is in maintenance mode focused on recovery. New deposits are not available.

3. Why do on-chain credit yields look higher than Aave?

Lenders take borrower credit risk that Aave’s overcollateralized loans avoid. The extra yield compensates for expected defaults.

4. What yield do I need to justify unsecured lending?

Calculate the break-even default rate: yield minus your alternative, divided by yield plus loss given default. A 12% pool with 60% losses needs annual default odds below about 10% to beat a 4.8% alternative.

5. Are curated vaults on Morpho or Euler the same as credit pools?

No, because they lend against on-chain collateral that a curator selects. Stream Finance showed that a curator’s collateral choices can still cause real lender losses.

References

The Block, Orthogonal Trading defaults on $36 million of loans on Maple Finance: https://www.theblock.co/post/192097/arummer@theblock.co

Blockworks, Maple Finance Opts for Immediate Defaults To Curb Bad Loans: https://www.blockworks.com/news/maple-finance-bad-loans-immediate-defaults

The Block, Maple Finance releases version 2.0: https://www.theblock.co/post/194907/maple-finance-releases-version-2-0-after-36-million-default-on-platform

This Week in DeFi, December 9: https://thisweekindefi.substack.com/p/this-week-in-defi-december-9

BlockCrunch, Inside the Maple MPL pool implosion: https://blockcrunch.substack.com/p/inside-the-maple-mpl-pool-implosion

War dieser Artikel hilfreich für Sie? Bitte teilen Sie uns in den Kommentaren unten mit, was Ihnen gefallen oder nicht gefallen hat.

About the Author: Chanuka Geekiyanage

Chanuka is a content writer, editor and contributor at Panaprium, specializing in long-form editorial articles. He conducts in-depth research and produces well-structured, carefully edited content with a strong focus on clarity, SEO best practices, and consistent formatting. His work spans lifestyle, fashion, wellness, and technical topics, and all articles are reviewed and refined to meet Panaprium’s editorial standards.

Wogegen Wir Kämpfen

Weltweit-Konzerne produzieren in den ärmsten Ländern im Übermaß billige Produkte.
Fabriken mit Sweatshop-ähnlichen Bedingungen, die die Arbeiter unterbezahlt.
Medienkonglomerate, die unethische, nicht nachhaltige Produkte bewerben.
Schlechte Akteure fördern durch unbewusstes Verhalten den übermäßigen Konsum.
- - - -
Zum Glück haben wir unsere Unterstützer, darunter auch Sie.
Panaprium wird von Lesern wie Ihnen finanziert, die sich unserer Mission anschließen möchten, die Welt völlig umweltfreundlich zu gestalten.

Wenn Sie können, unterstützen Sie uns bitte monatlich. Die Einrichtung dauert weniger als eine Minute und Sie werden jeden Monat einen großen Beitrag leisten. Danke schön.